Microcontrollers have become commonplace in the thirty years since their introduction. They have all but replaced mechanical and electromechanical components in the area of control over “real world” activities. For applications now controlled by microcontrollers, control functions therein are now much more functional, reliable, and economical.
Major improvements in microcontroller design since their introduction have made them nearly ubiquitous in modern control applications. The in-circuit emulator improved debugging and the integration of hardware and software. Embedded application development effectuated by C and other compilers has reduced software development time and allowed much larger programs and concomitantly more complex applications. One time programmability (OTP) of microcontrollers extended their utility, particularly for highly specialized and/or low volume applications. Programmability also improved the development cycle for users of microcontrollers.
Microcontrollers have embedded processors, memories, power sources, voltage references, voltage/power and temperature sensors, timers, oscillators, and other circuits. Various microcontrollers have differing features, including capacities. The 8-bit microcontrollers are an extremely useful, common, and well-populated class.
Contemporarily, there are thousands of different 8-bit microcontrollers from a number of sources. Nevertheless, selecting a microcontroller for a particular application and/or matching a particular microcontroller to a specific application remains a challenge. First, selecting a particular microcontroller from the many available can be confusing and tedious. After a selection is made, changing design requirements, engineering solutions, and/or unexpected higher capacity requirements often require scrapping the original selection and repeating the confusing and tedious selection process.
Conventionally, these problems may be addressed by custom designing a microcontroller with a “perfect,” e.g., exact, particular combination of required peripheral functionalities, and no surplusage, incorporating all needed functions, and eliminating a requirement for any external chips. This is demanding of time and resources, because it requires custom design and manufacturing operations for each selected application. It is expensive, in as much as it can take no advantage of the usual electronics industry economies of scale, which otherwise typically hold electronic prices at reasonably low levels.
Microcontrollers effectuate a wide range of applications in modem electronic installations into which they are functionally integrated. One major microcontroller utilization is the embedded system application. Most embedded system applications interface to the “real world.” This real world is analog in nature, and most microcontrollers interfacing with it offer an analog to digital (A/D) converter; true analog peripherals are rare. However, many microcontroller designs with real world interfacing embedded systems require that analog signals be multiplied, filtered, or otherwise conditioned before conversion to digital. While conventional analog functional components are available for use with microcontrollers, they are custom components and still require a separate microcontroller and an effective electrical coupling and signal synchronization and transfer modality to effectuate their use therewith. This is inefficient and costly.
Microcontrollers have a number of components to effectuate device application. Such components in conventional microcontrollers have fixed functions, which are disadvantageous in two major ways. First, in selecting a microcontroller for a particular application, it must be known in advance precisely which functions are required to effectuate that application and that this functional requirement is static. Second, specifying any particular function carries a cost, in as much as that function is static. The following example illustrates this second limitation.
A conventional microcontroller with “off the shelf” availability is selected for a particular application because it has a timer functionality, required by the application for which it is to be used. To effectuate this particular microcontroller's timer functionality, the microcontroller has two integrated timing components. However, the application at hand may be effectuated by the microcontroller if it had only a single timer component. This is wasteful of chip resources, power and computing demands, etc. Yet finding an exact, or even closer match from the finite supply of available microcontrollers with off the shelf availability is difficult and time consuming.
This limitation can be offset by negotiation with the microcontroller manufacturer for a custom designed and built chip, or the user, seeking the microcontroller for the particular application at hand may continue to search for another microcontroller with off the shelf availability, having components more closely matching the requirements of the application at hand. However, as discussed above, either of these solutions is also costly in terms of time, resources, and/or expense.
Further, microcontrollers employing conventional component technology have individual characteristic spectra of application, which are typically rather limited and static. Often, particular microcontrollers have rather precisely defined design functionalities, which are static and unchangeable, or changeable only in rather limited ways. Thus in this regard, conventional microcontrollers applications are inherently one dimensional and inflexible. This is also true of other circuits, such as an application specific integrated circuit (ASIC).
Conventional microcontrollers themselves are not reconfigurable to any convenient degree. A relatively small fraction of available conventional microcontrollers, and those implementing very general functions, have some degree of reconfigurability. However, the degree of reconfigurability is very limited. For example, one particular type of conventional microcontrollers implementing very general functions includes logic devices such as programmable gate arrays.
Programmable gate arrays typically are characterized by very fine grained logic architectures.
In so far as programmable gate arrays are reconfigurable at all, their reconfiguration is a static process, requiring a programmable gate array being so reconfigured to be out of service during the process, which takes an inordinate amount of time and requires a heavy price in computational resources. This is because the fine grain architecture of the programmable gate array being reconfigured demands thousands, for some common reconfigurations even millions of bits of information to be written, for each and every logic block requiring re-writing to effectuate the reconfiguration.
The conventional art is problematic because it generally fails to address the limitations of individual microcontroller and integrated circuit (IC) applicability and flexibility, and configurablity and programmability. Where reconfigurability is possible at all in conventional microcontrollers and ICs, it is typically achieved statically, with the microcontroller or IC out of service, to a very limited degree, and requires relatively long times and informational input to achieve. Custom designed analog-based devices are coupled with microcontrollers and/or ICs in such a way as to harmonize their operations in particular microcontroller/IC applications requiring analog functionality. Contemporary solutions to these problems using conventional resources are inadequate because of the time and effort required for custom choosing a particular conventional microcontroller/IC design for a certain application from a relatively limited field, resource costs of functionalities selected in the conventional microcontrollers/ICs selected, and the inordinate expense of custom chips, such as ASICs.